Journal of the American Medical Informatics Association : JAMIA
Jul 1, 2018
OBJECTIVE: To extract drug indications from a commercial drug knowledgebase and determine to what extent drug indications can discriminate between look-alike-sound-alike (LASA) drugs.
MOTIVATION: In biomedical research, chemical is an important class of entities, and chemical named entity recognition (NER) is an important task in the field of biomedical information extraction. However, most popular chemical NER methods are based o...
Journal of the Royal Society, Interface
Apr 1, 2018
Deep learning describes a class of machine learning algorithms that are capable of combining raw inputs into layers of intermediate features. These algorithms have recently shown impressive results across a variety of domains. Biology and medicine ar...
Advances in experimental medicine and biology
Jan 1, 2018
Medicine will experience many changes in the coming years because the so-called "medicine of the future" will be increasingly proactive, featuring four basic elements: predictive, personalized, preventive, and participatory. Drivers for these changes...
Studies in health technology and informatics
Jan 1, 2018
The field of medicine still reports errors because of insufficient knowledge or resources, work load or data not available at the right time and place, and this may be fatal for a patient. To improve the healthcare quality, a doctor needs accurate an...
Studies in health technology and informatics
Jan 1, 2018
We describe an approach to select semantically coherent specialty subsets based on the historical use of terminology by different service areas. Our approach uses rule-based and machine learning techniques to obtain a reduced set of 29 specialties.
SUMMARY: The miRBase database is the central and official repository for miRNAs and the current release is miRBase version 21.0. Name changes in different miRBase releases cause inconsistencies in miRNA names from version to version. When working wit...
IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
May 1, 2016
Discriminative appearance features are effective for recognizing actions in a fixed view, but may not generalize well to a new view. In this paper, we present two effective approaches to learn dictionaries for robust action recognition across views. ...